torcs-users Mailing List for TORCS - The Open Racing Car Simulator

Hi Toby

> I don't know whether I'm imagining it, but this version feels a bit 
> faster than 1.2.0 too. Could be compilation options but I'm not sure...

What makes a difference in my oppinion is if you compile specific for 
your processor. Before running ./configure I do on my PIII-550:
export CXXFLAGS="-march=i686"
export CFLAGS="-march=i686"
export CPPFLAGS="-march=i686"

> unintuitive when the going gets hairy - any recommendations?) I realise 
> that the wheels don't really slide out in the corners in quite the way 

The reasons are the characteristics and the setup of the cars. For 
racing cars the center of gravity is near the middle and the setup is 
optimized for neutral car behaviour. If you want your rear slide out you 
have to:
- brake or accelarate tough within the turn
- move the brake balance towards the rear
- reduce the downforce of the rear spoiler
- did you switch off ABS and ASR?
- are you racing on semi-pro or pro?

You can find the "human" driver setups in 
/usr/local/share/games/torcs/drivers/human if you want to change them.

> I'd expect. As I turn the car into the corner and tap the brakes to step 
> the back out, it's actually extremely difficult to develop a power slide 

That's like already mentioned just a matter of car setup, disabling the 
driving aids and the difficulty level .

> - the car slows down like it's running in thick treacle. I wonder if 
> this is due to the way that the tyres are modelled? Do they have the 
> same friction coefficients when sliding sideways as when braking in a 
> straight line? Is that something that can be altered (i.e. to simulate 
> thin tyres for example)? Can someone point me at the right parts of the 

Look in the TORCS sources below src/modules/simu/simuv2, wheel.cpp. For 
me it feels ok.

> source code for the car physics and even more importantly, how will 
> changing the physics models affect the robot drivers?

The robots have exactly the same physics like the human driver (it's a 
simulation, not a game;-)). You have just to know that the robots drive 
on the diffuculty level "semi pro" per default (if not different 
specified in the configuration file). I think all shipped robots are 
driving on semi-pro.

bye, bernhard.

-- 

visit my homepage http://n.ethz.ch/student/hwymann

Been knocking seconds off my fastest lap times with the new keyboard 
controls, which are nicely tuned and actually drive well considering 
that a keyboard is strictly digital... Good job!

I don't know whether I'm imagining it, but this version feels a bit 
faster than 1.2.0 too. Could be compilation options but I'm not sure...

Now that I can really seperate steering from everything else going on (I 
really must get a decent steering wheel - using a joystick is too 
unintuitive when the going gets hairy - any recommendations?) I realise 
that the wheels don't really slide out in the corners in quite the way 
I'd expect. As I turn the car into the corner and tap the brakes to step 
the back out, it's actually extremely difficult to develop a power slide 
- the car slows down like it's running in thick treacle. I wonder if 
this is due to the way that the tyres are modelled? Do they have the 
same friction coefficients when sliding sideways as when braking in a 
straight line? Is that something that can be altered (i.e. to simulate 
thin tyres for example)? Can someone point me at the right parts of the 
source code for the car physics and even more importantly, how will 
changing the physics models affect the robot drivers?

Thanks,
Toby Haynes

Hi all

today I've uploaded the robot tutorial chapter 6. As usual you can find 
it on http://n.ethz.ch/student/hwymann in the torcs section.

bye, have fun

bernhard.


-- 

visit my homepage http://n.ethz.ch/student/hwymann

Hi

> Hi all, it seems to me your game is really fine also considering it's a 
> freeware, I have a little trouble (win 2000) I can drive only with the 
 > mouse, what's about the keyboard that's not listed under 
OPTIONS/CONTROLS?

TORCS 1.2.0 has no keyboard support. In case you have vc++ 6.0 you can 
download the sources of "inofficial" 1.2.1 from http://torcs.org/tmp and 
compile them yourself, it includes keyboard support.

bye, bernhard.

-- 

visit my homepage http://n.ethz.ch/student/hwymann

Hi,

The keyboard driving is the main feature of the upcoming 1.2.1

The 1.2.1 should be available soon.

sergio simonetti wrote:
> Hi all, it seems to me your game is really fine also considering it's a 
> freeware, I have a little trouble (win 2000) I can drive only with the 
> mouse, what's about the keyboard that's not listed under OPTIONS/CONTROLS?
>  
> Thanks in advance.
>  
> Sergio Simonetti

Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
         AKA The Other Release Coming Soon (Skin'r)
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Thanks you, it will be commited to the CVS tonight.

Pajala Jussi wrote:
> Hi,
> 
> I found a little bug in the robottools, rttrack.cpp, function
> RtTrackLocal2Global.
> See below.
> 
> BR,
> 
> --Jussi
> 
> // robottools, rttrack.cpp
> 
> RtTrackLocal2Global(tTrkLocPos *p, tdble *X, tdble *Y, int flag)
> {
>     tdble CosA, SinA, r, a;
>     tdble tr;
>     
>     tTrackSeg *seg = p->seg;
>     switch (flag) {
>     case TR_TOMIDDLE:
> 	switch(seg->type){
> 	case TR_STR:
> 	    CosA = cos(seg->angle[TR_ZS]);
> 	    SinA = sin(seg->angle[TR_ZS]);
>             // must be divided by two to get middle of the track!
> // corrected line below
> 	    tr = p->toMiddle + seg->startWidth/2.0;
> // was:   tr = p->toMiddle + seg->startWidth;
> 	    *X = seg->vertex[TR_SR].x + p->toStart * CosA - tr * SinA;
> 	    *Y = seg->vertex[TR_SR].y + p->toStart * SinA + tr * CosA;
> 	    break;
> ...
> 
> 

Thanks,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
         AKA The Other Release Coming Soon (Skin'r)
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Hi all, it seems to me your game is really fine also considering it's a =
freeware, I have a little trouble (win 2000) I can drive only with the =
mouse, what's about the keyboard that's not listed under =
OPTIONS/CONTROLS?

Thanks in advance.

Sergio Simonetti

Hi,

I found a little bug in the robottools, rttrack.cpp, function
RtTrackLocal2Global.
See below.

BR,

--Jussi

// robottools, rttrack.cpp

RtTrackLocal2Global(tTrkLocPos *p, tdble *X, tdble *Y, int flag)
{
    tdble CosA, SinA, r, a;
    tdble tr;
    
    tTrackSeg *seg = p->seg;
    switch (flag) {
    case TR_TOMIDDLE:
	switch(seg->type){
	case TR_STR:
	    CosA = cos(seg->angle[TR_ZS]);
	    SinA = sin(seg->angle[TR_ZS]);
            // must be divided by two to get middle of the track!
// corrected line below
	    tr = p->toMiddle + seg->startWidth/2.0;
// was:   tr = p->toMiddle + seg->startWidth;
	    *X = seg->vertex[TR_SR].x + p->toStart * CosA - tr * SinA;
	    *Y = seg->vertex[TR_SR].y + p->toStart * SinA + tr * CosA;
	    break;
...

Am Fre, 2003-04-11 um 21.18 schrieb Darko K.:
> On Fri, 11 Apr 2003 21:22:07 +1200 Dietrich Radel wrote:
> 
> > 
> > Just tested the new Debian version in the directory
> > torcs.org/tmp/Debian. I get the error:
> > 
> > /usr/bin/torcs: 
> > line 50:  1267 Segmentation fault      ./torcs -l $LOCAL_CONF $*
> > 
> I compiled it on my Debian 3.0 and it works fine. Try doing the same
> and see if you get a stable version.

The packages are compiled and tested on my Debian/Sid. 
If I have the time, I will install woody and provide also a stable
version. But for now, you should compile it by yourself if it crashes...
Sorry.

Gernot

Dietrich Radel wrote:
> Hi guys,
> 
> Just tested the new Debian version in the directory torcs.org/tmp/Debian.
> I get the error:
> 
> /usr/bin/torcs: 
> line 50:  1267 Segmentation fault      ./torcs -l $LOCAL_CONF $*

Can you be more precise on the conditions ?

> 
> Keyboard seems to work. I think whether you use mouse or keyboard, on fast 
> computers, the steering is way way way to sensitive and near impossible to 
> drive in a straight line.

You can change the sensibility on the configuration screen.
For the mouse use, you can also increase the "dead zone" in order
to have more stability.

> 
> Oh yeah, and there was a car driving around with no body - just 4 wheels. LOL
> My wife quite enjoyed that. ;-)

What was the robot name of the "car" ?
May be it can be related to specific hardware/driver ?
What is your config (proc/video/mem...) ?

> 
> That being said, torcs is a great game though! Keep up the excellent work!

Thanks for the encouragements, they are appreciated :-)

> 
> Dietrich
> 
> 
> 

Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

On Fri, 11 Apr 2003 21:22:07 +1200 Dietrich Radel wrote:

> 
> Just tested the new Debian version in the directory
> torcs.org/tmp/Debian. I get the error:
> 
> /usr/bin/torcs: 
> line 50:  1267 Segmentation fault      ./torcs -l $LOCAL_CONF $*
> 
I compiled it on my Debian 3.0 and it works fine. Try doing the same
and see if you get a stable version.

> Keyboard seems to work. I think whether you use mouse or keyboard,
> on fast computers, the steering is way way way to sensitive and near
> impossible to drive in a straight line.
> 
It is much more enjoyable with a joystick or even a racing wheel!

Cheers,
         Darko

-- 

===============================================================
   "Freedom is just another word for nothing left to lose"
                            Janis Joplin, Me And Bobby McGee
===============================================================

Hi guys,

Just tested the new Debian version in the directory torcs.org/tmp/Debian.
I get the error:

/usr/bin/torcs: 
line 50:  1267 Segmentation fault      ./torcs -l $LOCAL_CONF $*

Keyboard seems to work. I think whether you use mouse or keyboard, on fast 
computers, the steering is way way way to sensitive and near impossible to 
drive in a straight line.

Oh yeah, and there was a car driving around with no body - just 4 wheels. LOL
My wife quite enjoyed that. ;-)

That being said, torcs is a great game though! Keep up the excellent work!

Dietrich

Hi,

You can try the js_demo program included in the PLIB example
package available on PLIB site.

A more basic test (test of the joystick driver) is available
here: <http://www.cnunix.com/ftp/linux/kernel/joystick/> it's
called jstest in the inpututils package.
another location is <ftp://atrey.karlin.mff.cuni.cz/pub/linux/input>

Another good start is the linux joystick driver page:
<http://atrey.karlin.mff.cuni.cz/~vojtech/joystick/>
the MS Gamepad is listed as supported :-)

I use the jscal program to calibrate my joystick.

Here is my joystick activation script:

############
modprobe evdev
modprobe joydev

if [ "$1" = "-c" ]
then
         jscal -c /dev/input/js0
         jscal -p /dev/input/js0 > /etc/joystick.cal
else
         sleep 1
         source /etc/joystick.cal
fi
############

This was very useful with my analog wheel, but I'm not sure
that calibration is of any interest with my USB wheel, expect testing
that the driver is loaded.

Just a question, are there some kind of joystick on the MS Gamepad
or are there just buttons ?

Bye,
Eric.

Dietrich Radel wrote:
> Hi there.
> 
> Here is the output from dmesg:
>   gameport0: Emu10k1 Gameport at 0xec00 size 8 speed 1217 kHz
>   input0: Microsoft SideWinder GamePad on gameport0.0 [3-bit id 75 data 5]
> 
> I already have /dev/input/js0  to js3. I also created a symlink called 
> /dev/js0 just incase.
> 
> Torcs still only allows configuration on my mouse. MS Gamepad (non-USB) still 
> not detected by torcs. My gameport is on my SoundBlaster Live! and has it's 
> own kernel support (emu10k1).
> 
> Any ideas,
> 
> Thanks
> Dietrich
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: ValueWeb: 
> Dedicated Hosting for just $79/mo with 500 GB of bandwidth! 
> No other company gives more support or power for your dedicated server
> http://click.atdmt.com/AFF/go/sdnxxaff00300020aff/direct/01/
> _______________________________________________
> Torcs-users mailing list
> Tor...@li...
> https://lists.sourceforge.net/lists/listinfo/torcs-users
> 
> 


-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Hi there.

Here is the output from dmesg:
  gameport0: Emu10k1 Gameport at 0xec00 size 8 speed 1217 kHz
  input0: Microsoft SideWinder GamePad on gameport0.0 [3-bit id 75 data 5]

I already have /dev/input/js0  to js3. I also created a symlink called 
/dev/js0 just incase.

Torcs still only allows configuration on my mouse. MS Gamepad (non-USB) still 
not detected by torcs. My gameport is on my SoundBlaster Live! and has it's 
own kernel support (emu10k1).

Any ideas,

Thanks
Dietrich

On Sat, 5 Apr 2003, D & E Radel wrote:

> > The total use of the CPU is a "normal" behaviour... The memory size
> > is partly due to the program itself, but mostly due to the Nvidia
> > drivers not releasing the memory every time. Your system must
> > be swapping that's why it's going slower.
>
> Ahh. We are noticing similar memory issues with a game we work on called
> Trackballs. Although the cpu is not at 99.3%, the memory is similar.

Sorry for being off-topic, to waste bandwith and your time,

Please don't see this advocacy on vendor, editor, mailer, browser, whatever
- AFAIK, the newest ATI cards also run only with a binary-only driver
and maybe I'll even by a NVidia card if I'm forced to. Sorry if I sound
angry and angainst NVidia.

I'm a religious Open Source User(using ATI Radeon Cards, Matrox MGA 400
and also tried a Voodoo3 for a short test) and was able to avoid NVidia
for the some time now.

I try to avoid any binary-only code whenever I can because it steals me the
freedom to do anything I want with my computer(as long as I know how).

Knowing-how is the main concern I have with any closed stuff because
if I don't know how to do it, I am dependent on some other party in
order to use my hardware, work on my documents or talk to others
over the internet.

Being independent of any other party is one of the main reasons
for using open source for me.

If I have to depend on some other party in order to do what I want with
my computer, I don't fully "own" my computer anymore and can only hope
the other party will continue to support me in the way I like and
not install any backdoors and digital rights management(DRM).

Maybe the DRM age will come anyway but I like freedom, open source
and indepence.

I know I have to make a compromise for example if I want to use a
winmodem under Linux, but fortunately I'm not dependent on a binary
only winmodem driver on order to connect to the internet, there are
free Ethernet and PCMCIA drivers to use with today's DSL lines.

End of religion, I won't mention it anymore and sorry for being off-topic.
Bernd

On Sat, 5 Apr 2003, D & E Radel wrote:

> > The screen sliding is a known problem with GLUT.
> > If you compile TORCS with freeglut this problem does not occur, but
> > you'll have to get the latest CVS archive of freeglut with the
> > modifications of Bernhard Kaindl in order to get your screen back
> > to the correct resolution after the game.
>
> Any chance of compiling binaries (Debian and other) against the freeglut?

It's not too hard to build if you know how: :-)

Install the packages libtool, autoconf and automake and
preferably checkinstall(for building a package from make install)

wget|curl http://freeglut.sourceforge.net/dist/current.tgz
rm -rf freeglut
tar xvfz current.tgz
cd freeglut/freeglut
libtoolize
(you can ingore a posssilbe  error about obsolete `AC_PROG_RANLIB')
aclocal
autoheader
(you can ingore a posssilbe  error about depreciated files)
automake -a
autoconf
./configure
make

Now run "checkinstall" if you have checkinstall to build a package
from make install or just run make install if you don't need
backups of possibly overwitten files, checkinstall would track
these file installs an build a package containing them:

/usr/local/bin/genfonts
/usr/local/bin/genstroke
/usr/local/bin/one
/usr/local/include/GL/freeglut.h
/usr/local/lib/libfreeglut-1.3.a
/usr/local/lib/libfreeglut-1.3.la
/usr/local/lib/libfreeglut-1.3.so
/usr/local/lib/libfreeglut-1.3.so.0
/usr/local/lib/libfreeglut-1.3.so.0.0.0

Go into the torcs-1.2.0 source directory of TORCS-1.2.0-src.tgz

and run this:

sed "s/-lglut/-lfreeglut-1.3/" Make-config >Make-config.new
diff Make-config  Make-config.new || mv Make-config.new Make-config

(diff should show you the change)

If you already compiled in this directory, run this:

rm src/linux/torcs
make

Now, the glut library should have a different name:

ldd src/linux/torcs | grep glut
        libfreeglut-1.3.so.0 => /usr/local/lib/libfreeglut-1.3.so.0 (0x40024000)

Install and try :-)

Bernd

PS:

If you have SuSE 8.1(8.2 includes it) and a fast connection, you can also get:
http://www.suse.de/~bk/Games/TORCS/8.1/torcs-1.2.0rc4-0.i586.rpm

If you have a different rpm-supporting distribution, you can get and try
rpm --rebuild with:
http://www.suse.de/~bk/Games/TORCS/8.1/SRPMS/torcs-1.2.0rc4-0.src.rpm

They are big downloads(20MB), but all important things except the berniw
drivers are included.

It's only RC4 so it misses the new very cool mouse sensivity configuration
screens but you can change the values it supports in
~/.torcs/drivers/human/preferences.xml:

    <attnum name="steer sensitivity" val="400"/>
    <attnum name="steer power" val="0"/>
    <attnum name="steer speed sensitivity" val="0.9"/>
    <attnum name="steer dead zone" unit="%" val="3.5"/>

(search for "speed" to find them)

Note you must (re)move ~/.torcs and /usr/share/games/torcs before the
install because the game will crash if anything(data/binraries/config)
is there which does not match this rc4 build!

> > Game is great, but for me there's a bug. One race is cool, but if move mouse
> > to far to left or right, the whole screen shifts like it's a virtual desktop
> > larger than screensize. My original desktop res is 1152x864 but I select
> > 640x480 in the game. Also, after the race everything jams up to a crawl.
>
> The screen sliding is a known problem with GLUT.
> If you compile TORCS with freeglut this problem does not occur, but
> you'll have to get the latest CVS archive of freeglut with the
> modifications of Bernhard Kaindl in order to get your screen back
> to the correct resolution after the game.

Any chance of compiling binaries (Debian and other) against the freeglut?

> > Buttons less responsive. I see in 'top' that Torcs is using 151MB of RAM and
> > 99.3% of CPU.  I downloaded the Debian packages. Anyone else have this
>
> The total use of the CPU is a "normal" behaviour... The memory size
> is partly due to the program itself, but mostly due to the Nvidia
> drivers not releasing the memory every time. Your system must
> be swapping that's why it's going slower.

Ahh. We are noticing similar memory issues with a game we work on called
Trackballs. Although the cpu is not at 99.3%, the memory is similar.

> > problem?
> >
> > PC:
> > AMD XP1800+ cpu
> > 256MB of DDR RAM (and plenty of swap space)
> > NVidia GF2 and 31.23 detonator drivers
> > Debian GNU/Linux "Sid"
> > XFree864.2
> > KDE3.1.1
> >
> > Your guys have done GREAT WORK so far! Keep it up!
>
> :-)
>
> >
> > Will Torcs be uploaded into official Debian "Sid" at any stage soon?
>
> Gerno ?
>
> >
> > Regards,
> > D.Radel.
> >
>
> Bye,
> Eric.
> --
> =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
>                         TORCS
>              The Open Racing Car Simulator
>                   http://torcs.org
> =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
>
>

On Fri, 4 Apr 2003, Bernhard Wymann wrote:
>
> > On my 1.8GHz P4m 1Gbyte laptop with Radeon 7800 FireGL card, the
> > performance is excellent. However - the overlays appear as greyscale and
> > aren't really readable. I'm assuming that this either a DRI/Mesa driver
> > bug or a reliance on something NVIDIA - has anyone else experienced
> > this? The laptop is running a fairly stock Redhat 8.0 distro.
>
> yes, on my brothers dell laptop it's the same (Radeon Mobility), i think
> it's a driver bug, because on other games the same problem occurs. for

I've had a strange drawing error with my Radeon Mobility with XFree86 4.2.0
which went away with 4.3.0.

It left an X Server lock when clipping or moving some 3D app windows which
could only be unlocked by killing the 3D app from remote which was fixed by
a patch which went into the CVS of dri-devel.sf.net at the beginning of March.

Bernd

hello,


Toby Haynes wrote:

>
> Graphics - the new CG Track 3 is an amazing piece of work. If 
> Christophe would PLEASE write a tutorial on track building I think 
> that would be appreciated by everyone :-)

thx

I'm writting a tutorial about creating tracks.
It will be  based on a new track called CG Track 4 ( not a surprise ;=) ).
I'm not sure if I would be able to write a tutorial as good as Bernhard 
Wymann's tutorial about bots.
It is rather long to write this tutorial, because creating tracks is not 
as easy as it could be,
and until now, I had planning issues.
But I hope the first chapter will be available in the next 2 weeks, and 
I feel really motivated
by all people enjoying the 1.2.0 version.

cg

Hi

> The driving stability appears to be better than it was before - the car 
> isn't quite so twitchy. I've only been playing on the easiest level so 
> far though so I'm not sure if there are extra stabilizers on the easy 
> levels :-)

yes, kind of. there are two parameters that change from rookie to pro, 
that's damage and adherance, so it will become more difficult with 
higher levels. i don't know what you mean with stability exactly, 
because for me the cars felt always right. you can't expect from normal 
cars "crusing" experience that it is as easy to deal with very high 
speeds and at the limit;-) look at real races how the cars shake and how 
the drivers are working to stay on the track...

> On my 1.8GHz P4m 1Gbyte laptop with Radeon 7800 FireGL card, the 
> performance is excellent. However - the overlays appear as greyscale and 
> aren't really readable. I'm assuming that this either a DRI/Mesa driver
> bug or a reliance on something NVIDIA - has anyone else experienced 
> this? The laptop is running a fairly stock Redhat 8.0 distro.

yes, on my brothers dell laptop it's the same (Radeon Mobility), i think 
it's a driver bug, because on other games the same problem occurs. for 
this billboards just standard blending is used, nothing vendor specific. 
you can disable the overlays (hit F1 for help during the race).

bye, bernhard.


-- 

visit my homepage http://n.ethz.ch/student/hwymann

I've installed TORCS 1.2 onto my two machines at home with different 
setups. Firstly, this is a great release - I particularly like the 
different player levels so I can get a feel for the cars prior to 
experimenting with damage, fuel, etc.
 
The driving stability appears to be better than it was before - the car 
isn't quite so twitchy. I've only been playing on the easiest level so 
far though so I'm not sure if there are extra stabilizers on the easy 
levels :-)
 
Graphics - the new CG Track 3 is an amazing piece of work. If Christophe 
would PLEASE write a tutorial on track building I think that would be 
appreciated by everyone :-)
 
Performance: On my Mandrake 9.1 128Mbyte Athlon 650MHz with a TNT2U 
card, TORCS is no longer really playable - the frame rate is around 
10fps - but everything displays correctly. I am probably suffering from 
slow stencil buffers, lack of memory or not having a graphics card with 
enough texturing units to do just one texture pass.
 
On my 1.8GHz P4m 1Gbyte laptop with Radeon 7800 FireGL card, the 
performance is excellent. However - the overlays appear as greyscale and 
aren't really readable. I'm assuming that this either a DRI/Mesa driver 
bug or a reliance on something NVIDIA - has anyone else experienced 
this? The laptop is running a fairly stock Redhat 8.0 distro.

Thanks for a great game.

Cheers,
Toby Haynes

On Fri, 2003-04-04 at 10:52, Dietrich Radel wrote:

> 
> Will Torcs be uploaded into official Debian "Sid" at any stage soon?

I do my best to become a debian maintainer soon;) It will last some time
before torcs will be a part of sid...

Gernot

Hi,

Dietrich Radel wrote:
> Hi. 
> 
> Game is great, but for me there's a bug. One race is cool, but if move mouse 
> to far to left or right, the whole screen shifts like it's a virtual desktop 
> larger than screensize. My original desktop res is 1152x864 but I select 
> 640x480 in the game. Also, after the race everything jams up to a crawl. 

The screen sliding is a known problem with GLUT.
If you compile TORCS with freeglut this problem does not occur, but
you'll have to get the latest CVS archive of freeglut with the
modifications of Bernhard Kaindl in order to get your screen back
to the correct resolution after the game.

> Buttons less responsive. I see in 'top' that Torcs is using 151MB of RAM and 
> 99.3% of CPU.  I downloaded the Debian packages. Anyone else have this 

The total use of the CPU is a "normal" behaviour... The memory size
is partly due to the program itself, but mostly due to the Nvidia
drivers not releasing the memory every time. Your system must
be swapping that's why it's going slower.

> problem?
> 
> PC:
> AMD XP1800+ cpu
> 256MB of DDR RAM (and plenty of swap space)
> NVidia GF2 and 31.23 detonator drivers
> Debian GNU/Linux "Sid"
> XFree864.2
> KDE3.1.1
> 
> Your guys have done GREAT WORK so far! Keep it up!

:-)

> 
> Will Torcs be uploaded into official Debian "Sid" at any stage soon?

Gerno ?

> 
> Regards,
> D.Radel.
> 

Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Hi,

I have searched in the Mesa 5.01 sources and gluBuild2DMipmaps was
defined (in src-glu/mipmap.c).
You can try the following actions:

1 - check the "checking for main in -lGLU... yes" line in the configure
     output

2 - Check that the /usr/X11R6/lib/libGLU.so.1 (or the one indicated by
     ldd /usr/local/share/games/torcs/torcs) contains
     the gluBuild2DMipmaps symbol (that I doubt)

Bye,
Eric.

Jaassu wrote:
> Hello !
> 
> It was my stupidity that libXxf86vm was missing on my Xfree4.3 installation,
> I had forgotten to install Xprog-package, sorry about that ;>).
> 
> Now compilation went without problems and I installed torcs and its datafiles.
> I was able to start torcs and configure its videomode, but when I tried to start practice session torcs crashes with this error:
> ....
> ........
> Saving /root/.torcs/config/raceman/practice.xml to /root/.torcs/config/raceman/practice.xml.old
> Saving /root/.torcs/config/raceman/quickrace.xml to /root/.torcs/config/raceman/quickrace.xml.old
> linuxModLoad: ...  modules/graphic/ssggraph.so: undefined symbol: gluBuild2DMipmaps
> /usr/local/games/bin/torcs: line 50: 29830 Segmentation fault      ./torcs -l $LOCAL_CONF $*
> sh-2.05b#
> 
> I have Mesa version 5.01 installed first in /usr-directory and NVIDIA-Linux-x86-1.0-4349 drivers after that.
> And I have deleted Xfree default include files in the /usr/X11R6/include/GL directory as mesa and nividia have
> installed their includes in /usr/include/GL
> 
> T Jari
> 


-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Hi. 

Game is great, but for me there's a bug. One race is cool, but if move mouse 
to far to left or right, the whole screen shifts like it's a virtual desktop 
larger than screensize. My original desktop res is 1152x864 but I select 
640x480 in the game. Also, after the race everything jams up to a crawl. 
Buttons less responsive. I see in 'top' that Torcs is using 151MB of RAM and 
99.3% of CPU.  I downloaded the Debian packages. Anyone else have this 
problem?

PC:
AMD XP1800+ cpu
256MB of DDR RAM (and plenty of swap space)
NVidia GF2 and 31.23 detonator drivers
Debian GNU/Linux "Sid"
XFree864.2
KDE3.1.1

Your guys have done GREAT WORK so far! Keep it up!

Will Torcs be uploaded into official Debian "Sid" at any stage soon?

Regards,
D.Radel.